Acting and Planning with Hierarchical Operational Models on a Mobile Robot: A Study with RAE+UPOM

2025-07-15

AI Summary

A semantic gap between symbolic planners and low-level robot control architectures undermines task execution robustness, particularly under perceptual noise and frequent action failures. Method: This paper introduces RAE+UPOM—a novel integrated system deployed for the first time on a mobile manipulator—featuring a shared hierarchical operational model that dynamically interleaves a reactive execution engine (RAE) with a UCT-based Monte Carlo online planner (UPOM). The architecture enables anytime planning and real-time responsiveness through hierarchical abstraction and online replanning. Contribution/Results: Experiments on object collection tasks demonstrate high success rates and exceptional robustness in realistic, noisy environments. The results highlight the critical role of tight coupling between reactive action execution and deliberative planning in enabling physical systems to adapt effectively to uncertainty and partial observability.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Multiagent Systems: Multiagent Planning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Robotic task execution faces challenges due to the inconsistency between symbolic planner models and the rich control structures actually running on the robot. In this paper, we present the first physical deployment of an integrated actor-planner system that shares hierarchical operational models for both acting and planning, interleaving the Reactive Acting Engine (RAE) with an anytime UCT-like Monte Carlo planner (UPOM). We implement RAE+UPOM on a mobile manipulator in a real-world deployment for an object collection task. Our experiments demonstrate robust task execution under action failures and sensor noise, and provide empirical insights into the interleaved acting-and-planning decision making process.

Problem

• Bridging symbolic planner models and robot control structures
• Integrating hierarchical models for acting and planning
• Ensuring robust task execution despite failures and noise

Innovation

• Hierarchical operational models for acting and planning
• Interleaving RAE with UCT-like Monte Carlo planner
• Robust execution under failures and sensor noise